1958] 



ANATOMY OF GUAYAXA MUTISIEAE PART II 



163 



the fact that the clearing techniques employed did not completely remove cell 

 contents, which thus stained in whole mounts and sections of cleared parts. The 

 taxa in which these were located with reasonable certainty are as follows : 



Gongylolepis huacliamacari subsp. neblinensis : cortex, pith, leaf 



G. b e nth ami an a : cortex, pith (fig. 4), involucral bracts 



G. fruticosa: cortex, leaf 



Duidaea pinifolia: cortex, leaf, achene 



Neblinaea promontorium : cortex, pith, peduncle 



Qaelchia eardonae : cortex, leaf, achene 



Although laticiferous cells may occur more widely in the plant body than is indi- 

 cated for the above taxa, their presence in the genera listed is of considerable 

 interest. "With the exception of Achnopogon, in which their presence could not 

 be definitely established, laticiferous cells are present in all the Guayana genera 

 of the Gerberinae. The presence, in addition, of laticiferous cells in Quelchia is 

 noteworthy in view of its systematic position, and it may be that Quelchia is 

 closer to Gerberinae than Stenopadus, Stomatochaeta, Chimantaea, and Glns- 

 sarion, which apparently do not possess laticiferous cells. 



Beside the interest inherent in a new record of laticifers for the tribe Mutisi- 

 eae, the fact that they are of the type found by Col in Arctotidae, Cynareae, 

 and Vernonieae is of possible significance. Col,, however, records anastamosing 

 articulated laticifers in addition in Gazama of Arctotidae and in the subtribe 

 Carlinae of Cynareae. The writer considers, however, that Col's designation 

 (p. 169) of individual laticiferous cells as 'formes imparfaites .... temoinage 

 d'une .evolution extensive on regressive" is open to doubt. Tt may be that in 

 such individual laticiferous cells, singly or in groups, we are witnessing a primi- 

 tive, rather than a degenerate form of tin 1 well-developed laticifers character- 

 istic of Cichorieae. 



Secretory canals. The term "secretory canal'' here refers to the structures 

 sometimes designated as "oil canals" or " resin canals" by various authors. 

 Where present, they are always an intercellular formation, thus differing from 

 laticifers, which are cellular in nature. AVithin the genera of this study no true 

 secretory canals — in the sense of their discrete structure in Heliantheae — can 

 be said to occur. In parenchyma or sclerenchyma of cortex and pith in Sti no- 

 p&dus and Stomatochaeta (figs. 1, 2), however, resins are abundantly secreted 

 into intercellular spaces; these resins carbonize, giving a blackish color to the 

 cortex and pith of these genera. Although such resin secretion is present even 

 in leaves, involucral bracts, and achenes of these genera, in no case could a 

 definite secretory canal be said to be present. It is of interest that secretion of 

 resins into vessels and libriform fibers of the secondary xylem is also character- 

 istic of these and other Gochnatinae (Carlquist 1957b). 



By contrast, none of the Gerberinae studied showed this type of resin secre- 

 tion. It may be that secretion of resinous materials into the laticiferous cells 

 replaces deposition in intercellular spaces in these genera. 



True secretory canals occur in some, although they are absent in many genera 

 of Mutisieae. Van Tieghem (1872) has reported them in the root of Stifftia 

 chrysantha, Gerhera schimperi, and others. He notes their presence in stem and 

 leaf of Ainsliaea acerifolia, Hyalis spartiodes, Lycoseris mexicana, Mutisia retusa. 

 Nassauvia cligitata, and Polyachrus villosus. Recently, Cabrera (1951) has dem- 

 onstrated their presence in Aphyllocladus denticulatus, A. ephedroides, and 

 Plazia. It is of interest, therefore, that secretory canals occur in all subtribes 



